Gamma-Ray Observations with CALET: Exposure
Map, Response Functions, and Simulated Results
Nicholas Cannady∗†and
Department of Physics and Astronomy, Louisiana State University, Baton Rouge, LA, 70803,
USA
E-mail: [email protected]
Michael L. Cherry for the CALET Collaboration
Department of Physics and Astronomy, Louisiana State University, Baton Rouge, LA, 70803,
USA
E-mail: [email protected]
The CALorimetric Electron Telescope (CALET) is a space-borne cosmic ray instrument planned
for installation on the JEM-EF platform on the International Space Station (ISS) in 2015. The
CALET collaboration is a Japan-led international team involving researchers in Italy and the U.S.
In addition to precise measurement of the cosmic ray electron and nuclei spectra, the CALET
calorimeter will be capable of gamma-ray observations in the energy range 10 GeV - 10 TeV.
This paper presents a study of the expected gamma-ray signal measured by CALET in the first
year on orbit. The ISS zenith pointing is simulated at a time resolution of 1 second in order to
estimate the exposure map on the sky. The instrument response functions and simulated results of
gamma-ray/electron separation for the calorimeter are discussed and used to estimate the expected
point source and galactic diffuse signals in the energy range 10 GeV - 500 GeV based on known
fluxes measured by Fermi-LAT.
The 34th International Cosmic Ray Conference,
30 July- 6 August, 2015
The Hague, The Netherlands
∗ Speaker.
† CALET
supported in the USA by NASA grant #NNX11AE01G
c Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike Licence.
http://pos.sissa.it/
Nicholas Cannady
Gamma-Ray Observations with CALET
1. Introduction
The CALorimetric Electron Telescope (CALET), a Japanese-led cosmic-ray experiment with
Italian and US collaborators, is scheduled for deployment on the International Space Station (ISS)
in the middle of 2015 for high-accuracy measurement of electrons above 10 GeV and up to and
beyond 10 TeV, as well as protons, heavier nuclei, and gamma-rays [1].
This paper examines the expected gamma-ray signal in CALET after 1 year of on-orbit data
collection. An algorithm is developed and evaluated for the separation of a gamma-ray event dataset
from the electron dataset based on simulations performed with the EPICS and Cosmos Monte Carlo
code [2]. An exposure map is generated for the 1-year CALET orbit on the ISS and presented along
with the resulting expectations of gamma-ray counts. Source fluxes and background modelling are
based upon recent Fermi-LAT results.
2. The CALET calorimeter
The CALET CALorimeter (CAL) is the primary detector on the CALET mission. It is
composed of three main components: the CHarge Detector (CHD), the IMaging Calorimeter
(IMC), and the Total AbSorption Calorimeter (TASC). The CHD is composed of two crossed layers
of plastic scintillator strips with the purpose of measuring the ionization energy deposit of primary
particles entering the CAL. The IMC is eight layers, each made up of a pair of crossed layers of
1mm2 plastic scintillator fibers. Interspersed between the eight layers are thin sheets of tungsten.
With a total depth of 3 X0 for normal incidence, the IMC will be capable of detailed measurement
of the early shower development for high-precision reconstruction of the primary particle trajectory.
The TASC contains 12 layers of PWO logs alternating in xy-orientation (total thickness 27 X0 for
normal incidence), which will absorb the majority of the energy deposited by particle showers [1].
A schematic of the CAL is shown in Figure 1 without any of the shielding that is present in the
Monte Carlo simulations.
Figure 1: Simulation of a 1 TeV electron in the CAL.
2
Nicholas Cannady
Gamma-Ray Observations with CALET
3. Gamma-ray/electron separation
Distinguishing between primary electrons and photons will be necessary for any CALET
gamma-ray science goals. The algorithm developed and described in this section is based only
on the signal in those CHD strips which were directly passed through by the primary particle,
henceforth referred to as "hit" components. The trajectories themselves are not reconstructed, but
taken directly from the Monte Carlo truth. The data used for the following results are generated by
the EPICS and Cosmos Monte Carlo simulation package on the LSU High Performance Computing
SuperMike-II cluster.
Gamma-ray and electron events were generated isotropically on a partial sphere (θ ≤ 110◦ )
with an E−1 spectrum in three energy bins: 10-100 GeV, 100-1000 GeV, 1-10 TeV. Events were
generated for each species in each of these energy bins and subsequently filtered based on their
trajectories. The requirements placed on events for analysis follow the "type B" geometry as
defined in [3]. Explicitly, this enforces that the primary particle passes through both the top of
the CHD and the bottom of the TASC.
For each of the type B events, the energy deposits in the hit strips in CHDx and CHDy were
summed, respectively. Histograms of these deposits (see Figure 2) verify that the distribution
for electrons appears as a minimum ionizing particle (MIP) distribution, whereas the gamma-rays
primarily exhibit zero energy deposit. Some gamma-rays pair produce in the aluminum structure
above the detector or in the CHD itself, leading to partial energy deposits due to secondary electrons
and a feature around 4 MeV (2 MIP) clearly seen in the 10-100 GeV and 100-1000 GeV histograms.
The pair production cross section and backscatter pollution of the CHD signal increase with photon
energy, leading to an increasing fraction of non-zero energy deposits in the CHD at higher energies.
10-100 GeV electrons
100-1000 GeV electrons
1-10 TeV electrons
10-100 GeV gamma-rays
100-1000 GeV gamma-rays
1-10 TeV gamma-rays
Figure 2: Histograms of CHD energy deposits in each energy bin, summed over hit strips in the layer.
3
Nicholas Cannady
Gamma-Ray Observations with CALET
Figure 3: Energy deposits in CHDy vs. CHDx for both electrons and gammas in each energy range. The
blue lines indicate the cuts made to separate the samples.
In the 1-10 TeV histogram, the 4 MeV feature is indistinguishable from the continuum due to the
large fraction of interacting gamma-rays and energy deposits from backscattered particles.
Scatter plots showing the total energy deposited in the hit components of CHDy as a function
of the total energy deposited in the hit components of CHDx provide a graphical representation
of the cuts used to separate the gamma-rays from the electrons. As can be seen in Figure 3, two
straight lines (one nearly horizontal, the other nearly vertical) can be drawn which largely divide
the samples. The functional requirement for identifying a gamma-ray by this scheme can be written
CHDx_dE ≤ 0.09 × (CHDy_dE + 0.01)
CHDx_dE ≥ (CHDy_dE/0.09) − 0.01
or
(3.1)
Table 1: Efficiency for separation of electrons and gamma-rays in type B geometry into individual datasets
Species
e-
γ
Energy
10-100 GeV
100-1000 GeV
1-10 TeV
10-100 GeV
100-1000 GeV
1-10 TeV
# in-geom.
7675
7554
7485
7482
7649
7573
# ID as γ
8
5
5
7217
7275
5797
% ID as γ
0.104%
0.0662%
0.0668%
96.5%
95.1%
76.5%
# ID as e
7667
7549
7480
265
374
1776
% ID as e
99.9%
99.9%
99.9%
3.54%
4.89%
23.5%
The criterion in Eq. 3.1 was applied to all of the type B events in order to characterize the
loss of gamma-rays from the sample and the rejection of electrons. The statistical results are
summarised in Table 1.
It should be emphasized that these results are as of yet preliminary, and only encompass events
of type A or B. Further work is in progress to extend this study to type C and D events using deposits
in the IMC as well. Using the IMC signals, tracking of the primary particle can be determined
instead of using the Monte Carlo truth. This trajectory and the associated error can be used to
identify those IMC fibers which could have been passed through by the primary particle. The
energy deposits in these fibers will show an ionization deposit if the primary particle is an electron,
which can be used to further improve the electron rejection power. These calculations will also
provide good estimates of the effect of trajectory reconstruction error on the ability to make a clean
separation.
4
Nicholas Cannady
Gamma-Ray Observations with CALET
4. CALET 1-year exposure map
When deployed on the ISS, CALET will always be oriented toward the zenith [3]. It has a
field of view radius and a high orbital inclination, and will, as such, be an all-sky monitor of
high-energy gamma-rays. In order to estimate the number of photons from known sources (diffuse
galactic emission, isotropic emission, and persistent point sources), a map of exposure time on the
sky has been generated. The resolution of the sky map was chosen to match the Fermi-LAT galactic
diffuse background model [4], which is binned to 0.125◦ × 0.125◦ in galactic coordinates. Orbital
ephemeris information was generated for the ISS orbit for a full year at 1s resolution using the AGI
Satellite Toolkit (STK) software. The calculated pointing directions were transformed to galactic
coordinates and used to determine the total duration for which each bin on the sky would be within
the field of view of the instrument. The resulting exposure map is shown in Figure 4.
45◦
Figure 4: CALET exposure map for continuous observation from Nov. 1, 2015 through Oct. 31, 2016.
Sharp features are due to the not yet accounted for zenith angle dependence of the instrument response.
5. Estimated number of events
The fluxes given in the Fermi-LAT models have units of [Φ] = photons cm−2 s−1 sr−1 MeV −1 .
To calculate the number of photons incident on the instrument, the following expression was used
N = Φ × Ae f f × texp × Ωbin × ∆Ebin
(5.1)
The effective area Ae f f is given as a function of event geometry in Figure 3 of [3]. The
value for type B geometry is not available, so the value of 500 cm2 for the more restrictive type
A geometry is used to provide a lower bound. The exposure time texp is taken for each bin from
the generated exposure map in Figure 4. The solid angle of the each sky bin can be calculated as a
function of the bin center galactic latitude.
The energy bin sizes (∆Ebin ) are logarithmically spaced and vary based on the Fermi-LAT model
being considered.
5
Nicholas Cannady
Gamma-Ray Observations with CALET
5.1 Diffuse galactic and isotropic
The Fermi-LAT background models [4] give fluxes for the isotropic and galactic diffuse
gamma-ray signals. Since the energies are binned differently for the two models and the
background spectra are not within the scope of this paper, the contributions from all energy bins
above 10 GeV (reaching up to 500 GeV for galactic sources and up to 600 GeV for the isotropic
signal) are summed to obtain a total number of background photons in each sky bin. Summing
the contributions from every sky bin, we calculate that CALET will be exposed to more than 3500
in-geometry photons at energies ≥ 10 GeV from the known diffuse flux per year.
5.2 Persistent sources
The estimate of counts above 10 GeV from point sources was taken from the first FermiLAT catalog of high-energy gamma-ray sources [5]. Extended sources which subtend less than
2◦ are included in this catalog, while larger extended sources are contained in the diffuse galactic
background model [4]. The total number of photons from these sources is calculated to be on the
order of 350. A list of the six brightest sources for CALET is provided in Table 2. All sources with
more than three incident photons are also shown on a sky map in Figure 5.
Table 2: The six brightest 1FHL sources detected in one year of CALET orbit.
1FHL designation
J0835.3-4510
J0534.5+2201
J2028.6+4110e
J1104.4+3812
J0617.2+2234e
J0633.9+1746
Common ID
Vela
Crab
MGRO J2031+41
Mkn 421
IC443
Geminga
Type
PWN
PWN
N/A
HBL
Shell
PSR
# of photons
22
19
12
11
8
7
Note that the sources whose 1FHL designation ends with an ’e’ are extended sources. The
photons from these sources will be spread across several sky bins. The background in each
individual sky bin was calculated, considering both the diffuse gamma-ray emission and the
contamination from electrons (flux taken as upper limit of PAMELA results for the 10-600 GeV
range [7], then reduced by the appropriate electron rejection factor), and found to be negligible
compared to the signal from these sources.
The estimated number of gamma-rays in table 2 reflect the expected CALET energy threshold
of 10 GeV. Similar estimates have been previously made if this threshold were lowered to 1 GeV
[6] for Vela (˜2100), Geminga (˜1200), and the Crab (˜360). Including the flux in the 1 GeV - 10
GeV energy band in the calculations in this paper for these sources, we estimate the number of
photons to be ˜2500 for Vela, ˜1500 for Geminga, and ˜380 for the Crab. The small discrepancy
between these numbers and those in [6] result from small differences in the effective area used for
each study. It is calculated in [6] that lowering the energy threshold from 10 GeV to 5 GeV would
result in a factor of 5-10 increase in the number of detected gamma-rays from persistent sources.
6
Nicholas Cannady
Gamma-Ray Observations with CALET
Figure 5: 1-year count map for CALET gamma-rays. The background color represents intensity of diffuse
emission in the area. The stars are steady point sources, with the size of the marker proportional to the
number of photons. Only sources with ≥ 3 photons are shown on the plot.
6. Observation of transient systems
A significant strength of CALET for high-energy gamma-rays is the large orbital inclination
of the ISS and the large field of view. These properties make CALET very well-suited as an all-sky
monitor for high-energy flaring activity. It is possible, as mentioned in [6], that the energy threshold
will be lowered to 1-5 GeV rather than the 10 GeV used in this paper. Not only would a decrease
in threshold to 5 GeV increase the number of detected photons from persistent point sources by a
factor of 5-10, it would also improve CALET’s performance as a monitor for transient systems.
As an example, the expected number of photons was calculated for two historical high-energy
Crab flares. The flares will be characterized by their duration (τ) and the ratio of the average flux
during the flare to the ambient flux (α). For the April 2011 flare, τ ≈ 10 days and α ≈ 30. [8], and
for the March 2013 flare, τ ≈ 20 [9]. If a flare were to occur on a day where CALET could ideally
observe it, the exposure time is approximately 12600 seconds (25% of the full day). The expected
signal from an α = 25 flare is then N ≈ 2 photons. For the historical flares, the fractional exposure
time is taken to be 16% of the total flare time, which is the ratio of the exposure time calculated for
the Crab in the orbit simulation to the total time in one year. These parameters yield an estimate of
15 photons over the full flaring period for either flare. This translates to a 3.8σ excess above the
ambient background and typical Crab signals.
A confirmed observation of the Crab in the CALET energy range would indicate an inverse
Compton scattering component in the flare mechanism. All previous Crab Nebula flares have been
seen in the range 20-200 MeV with spectra consistent with synchrotron emission. Observation of
transient activity at CALET energies would yield unique information about the flaring mechanism.
7
Nicholas Cannady
Gamma-Ray Observations with CALET
7. Summary
The CALET calorimeter will perform very well in the identification of a gamma-ray dataset,
including rejection of contaminating electrons at better than 1 in 104 . This rejection factor
is expected to improve further with the inclusion of IMC data in the tracking and separation
algorithms. CALET will be able to observe bright, steady gamma-ray sources, with sensitivity
up to TeV energies. Furthermore, CALET will be a very useful instrument for all-sky detection
and observation of flares at ≥ 100 GeV energies from transient systems due to its large field of
view and high orbital inclination aboard the ISS.
Acknowledgements
The CALET experiment is supported by NASA in the United States, JAXA in Japan, and ASI in Italy. Portions
of this research were conducted with high performance computing resources provided by Louisiana State University
(http://www.hpc.lsu.edu). AGI Satellite Toolkit (STK) software was used in the generation of some results in this paper.
The authors also appreciate the efforts of the Fermi-LAT team in development of the background models used. Reference
was made to the University of Chicago TeVCat catalog for collections of source-related information. Financial support
for the lead author partially provided by a Louisiana State University Board of Regents Graduate Fellowship.
References
[1] Shoji Torii for the CALET Collaboration, The Calorimetric Electron Telescope (CALET) for High
Energy Astroparticle Physics on the International Space Station, Proc. 33rd ICRC (Rio de Janeiro),
2013, and these proceedings.
[2] K. Kasahara, Home Page for EPICS and Cosmos, [http://cosmos.n.kanagawa-u.ac.jp/]
[3] Masaki Mori for the CALET Collaboration, Expected Performance of CALET as a High Energy
Gamma Ray Observatory, Proc. 33rd ICRC (Rio de Janeiro), 2013.
[4] Fermi Science Support Center, Galactic Interstellar Emission Model and Isotropic Spectral Template,
LAT Background Models,
[http://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html]
[5] M. Ackermann et al., The First Fermi-LAT Catalog of Sources above 10 GeV, ApJS, 209, 34
[astro-ph.IM/1306.6772v3], 2013.
[6] Alex Moiseev for the CALET Collaboration, CALET Perspectives in High-Energy Gamma-Ray
Observations, Proc. 33rd ICRC (Rio de Janeiro), 2013.
[7] O. Adriani et al., The cosmic-ray electron flux measured by the PAMELA experiment between 1 and
625 GeV, Phys. Rev. Lett., 106, 20 [astro-ph.HE/1103/2880v1], 2011.
[8] R. Buehler et al., Gamma-ray activity in the Crab Nebula: the exceptional flare of 2011 April, Ap. J.,
749, 26 2012
[9] M. Mayer et al., Rapid gamma-ray flux variability during the 2013 March Crab Nebula flare, Ap. J.
L., 775, L37 2013
8